sign in sign up
<<
Home , Flat Hill, Hawkwatching, Hawk Mountain, Scotts Mountain

Using Autumn Watch to Track Raptor Migration and to Monitor Populations of North American Birds of Prey1

Kyle McCarty2 and Keith L. Bildstein2 ______Abstract Raptors are secretive, area-sensitive predators whose (Bildstein 1998). One particularly cost-effective populations can be logistically difficult and financially method for monitoring populations of these birds is to prohibitive to monitor. Many North American popula- sample regional and even continental populations at tions of raptors are migratory however, and on mig- traditional migratory bottlenecks and concentration ration raptors are frequently counted at traditional points (Zalles and Bildstein 2000). An incipient net- migration watchsites. Experiences at Hawk Mountain work of raptor migration watchsites (i.e., lookout Sanctuary (HMS) and elsewhere suggest that long-term points typically situated on mountain-tops, coastal migration counts can be used to monitor regional plains, river valleys, and lakesides and other migration populations of raptors. Hawkwatchers have collected “leading lines”) exists in North America, and during count data on standardized Hawk Migration of North the past 70 years migrants have been counted on a America (HMANA) Daily Report Forms since the mid- regular or irregular basis at more than 1,800 short-term 1970s. In 1998 HMS, HMANA, the National Audubon or permanent watchsites (McCarty et al. 2000). Society, and the Lab of Ornithology at Cornell Uni- versity began Internet-based data entry of migration Although most raptor migration watchsites are in the watchsite counts at the BirdSource website. By autumn northeastern United States, the recent addition of 2002, the Autumn Hawk Watch web page was col- season-long migration counts at sites in Gulf Coast lecting and displaying daily count reports from 66 Texas (Smith et al 2001) and Mexico (Zalles and watchsites in three Canadian provinces, 26 United Bildstein 2000), and in Caribbean slope Costa Rica States, and in Mexico, Costa Rica, and Bolivia. Autumn (Bildstein, pers. obs.), together with the establishment Hawk Watch provides participants and interested of an array of watchsites in the western United States parties with near real-time maps and tables that in the 1980s (Smith and Hoffman 2000), has created a document the movements of raptor migration across potentially effective network for assessing regional and the Americas each autumn. The web page also captures continental populations of migratory birds of prey. count data for later use in monitoring raptor pop- Since the mid 1970s, most migration data have been ulations, and provides HMANA with timely summaries collected using Hawk Migration Association of North of each count, which are published in the HMANA America (HMANA) Daily Report Forms, or “green Journal of Hawk Migration Studies. sheets.” As such, much of the information collected at these sites is compatible and regional and continental assessments of population change are possible. Unfor- tunately, most of this information is not available in Key Words: Autumn Hawk Watch; Hawk Migration electronic format in a single database, making broad- Association of North America; Hawk Mountain scale, multi-site analyses difficult. However, monitor- Sanctuary; Internet; migration monitoring; raptors. ing numbers of migrating at single migration watchsites has been used to great effect (e.g., Carson 1962, Newton 1979, Kerlinger 1989, Bednarz et al 1990, Zalles and Bildstein 2000, Mueller et al. 2001), and the use of data from networks of sites holds great Introduction promise. The establishment of several North American- style, season-long count efforts in Mesoamerica and Raptors are wide-ranging and often secretive predatory the Caribbean Basin in the past 10 years, for example, birds, many of whose populations are logistically dif- increases the potential for tracking world populations ficult and financially prohibitive to survey and monitor of three species, Mississippi Kite (Ictinia mississip- ______piensis), Broad-winged Hawk (Buteo platypterus), and Swainson’s Hawk (B. swainsoni), as well as conti- 1 A version of this paper was presented at the Third Interna- nental populations of a fourth species, Swallow-tailed tional Partners in Flight Conference, March 20-24, 2002, Kite (Elanoides forficatus) (Zalles and Bildstein 2000, Asilomar Conference Grounds, California. 2Hawk Mountain Sanctuary Acopian Center, 410 Summer Bildstein and Zalles 2001). Valley Road, Orwigsburg, 17961. E-mail: [email protected].

USDA Forest Service Gen. Tech. Rep. PSW-GTR-191. 2005 718 Autumn Hawk Watch - McCarty and Bildstein

Here we (1) report several examples of how a single- States in the early 1970s. Even more remarkable is that site, long-term migration watchsite database has been shifts in overall numbers of Bald Eagles seen at the site used to track changes in raptor populations and mig- lagged declines and subsequent increases in the ratios ration behavior, and (2) describe a new webpage that of juvenile-to-adult birds by 5 to 10 years, exactly as uses the power of the Internet to capture, analyze, and expected if the shifts in ratios of age classes reflected display the results of counts at a network of migration shifts in reproductive success in the monitored pop- watchsites across North and Central America. ulation (Bildstein 1998).

A more recent example of the utility of the HMS database involves its use, together with counts from The Conservation Potential of other migration watchsites and information from add- Migration Watchsite Count Data itional geographically explicit databases. Beginning in the 1980s and extending into the early 1990s, many (HMS) was founded in migration watchsites in the northeastern United States 1934 by conservationist Rosalie Edge to stop the began reporting substantial declines in numbers of shooting of thousands of migrating raptors along the Sharp-shinned Hawks (Accipiter striatus). The declines Kittatinny Ridge in the central were especially notable at coastal watchsites, where of eastern Pennsylvania, 120 km northwest of Phila- flights consisted largely of juvenile individuals. Al- delphia, Pennsylvania. In late September 1934, Hawk though initial reports of the decline were accompanied Mountain’s Maurice Broun began counting migrating by suggestions of natural population cycling or shifts in raptors from what was then called Observation Rocks. wintering range, by the early 1990s additional reports Although Broun’s counts were initiated primarily to became more ominous, and included suggested links to document the numbers of raptors being “saved” at the widespread habitat or prey-base loss induced by acid Sanctuary, so as to enlist financial support for the precipitation or declines in populations of Neotropical conservation effort, it quickly became apparent that a songbird migrants (Viverette et al. 1996). series of annual counts would enable conservationists to monitor regional populations of birds of prey. To- To test the hypothesis that the declines resulted from day, Hawk Mountain maintains the longest and most northward shifts in wintering areas of the regional complete record of raptor migration in the world population, researchers at Hawk Mountain compared (Bildstein and Compton 2000). migration count data from the Sanctuary and Cape May Point, with those from National Audubon One of the first conservation uses of the Hawk Moun- Society Christmas Bird Counts (CBCs) north and south tain long-term database was an analysis of annual of the two migration watchsites. The analysis of CBCs ratios of juvenile and adult Bald Eagles (Haliaeetus conducted from 1979 through 1989 revealed that num- leucocephalus) seen at Hawk Mountain before and th bers of Sharp-shinned Hawks seen on CBCs north of after the widespread use of DDT in mid-20 Century the two migration count sites increased substantially North America. A bimodal pattern in the seasonal whereas numbers seen on CBCs south of the two timing of the flight of Bald Eagles at Hawk Mountain, watchsites had not. The results support the shift-in- with a major movement in late August-early Septem- migration behavior hypothesis (i.e., migratory short- ber, and a second smaller movement peaking in mid- stopping). That Cape May Point counts declined more November, suggested that although both “southern” abruptly than those at Hawk Mountain may reflect the and “northern” birds migrated at the site, most of the fact that the Sharp-shinned Hawk flight at the former birds were southern Bald Eagles nesting in Florida. In site consists almost entirely of juvenile individuals, 1952, a concerned Maurice Broun began commenting whereas that at the latter includes many more adults. on a substantial decline in the ratio of juvenile-to-adult Juveniles are more likely to modify their migration Bald Eagles at the site, which began in the late 1940s patterns in light of changed environmental conditions and, thereafter, remained low through the 1950s, than are adults (Berthold 1993). 1960s, and mid-1970s (Bildstein 1998). A decade after Broun reported the shift, Rachel Carson used the same Although the reason for the shift in behavior remains database in Silent Spring to support her arguments for uncertain, the change coincided with a series of part- the impact of organochlorine pesticides on populations icularly mild winters, as well as with increases in the of Bald Eagles and other species of predatory birds numbers of backyard birdfeeders in northeastern North (Carson 1962). America. Sharp-shinned Hawks rank above domestic cats (Felis catus) as the number one predator of birds at Particularly notable is that the same database, which st backyard birdfeeders (Dunn and Tessaglia 1994), and now extends into the 21 Century, was also used to the growing number of bird feeders in the northeastern track the eventual recovery of populations United States together with milder winters may have following bans on DDT in Canada and the United combined to produce the shift (Viverette et al. 1996).

USDA Forest Service Gen. Tech. Rep. PSW-GTR-191. 2005 719 Autumn Hawk Watch - McCarty and Bildstein

This latter example suggests the utility of migration continent report their results to HMANA each year, counts in documenting shifts in migratory behavior, as contributing this information to a growing archive of well as in documenting changes in overall population, more than 80,000 standardized daily report forms and also highlights the importance of using count data (DRFs). Most of the counts are reported in the in conjunction with other geographically explicit data- HMANA journal every year (fig. 1). bases such as the Christmas Bird Count. In 1996, HMS partnered with HMANA and began cataloging and archiving the DRFs. In an effort to make the data more available to potential users, as well Realizing the Conservation Potential of as to speed the summarization and analyses of the data, Migration Watchsite Count Data Using in 1998 HMANA, HMS, the National Audubon Soc- Autumn Hawk Watch iety (NAS), and the Laboratory of Ornithology at Cornell University (CLO) created the International HMANA was founded in 1974 when it became clear Broad-winged Hawk Survey. The Survey enabled (1) that large numbers of individuals were starting to hawkwatchers from North to South America to enter count raptors on migration, (2) that it was important to counts of migrating Broad-winged Hawks into a web- organize the wealth of information derived from those based data set and to view their collective results in counts, and (3) that there was a need to develop a near real-time animated maps of “waves” of these standard field sheet to collect these data. The plan was raptors migrating across North America. In 1999 the to distribute count data to the public and the scientific web page (now called Autumn Hawk Watch) was community, and to improve techniques to monitor expanded to include all species of diurnal raptors populations of birds of prey (Roberts 2001). Today, migrating in North America, as well as information on cooperators at several hundred watchsites across the weather conditions.

Figure 1—The Hawk Migration Association of North America daily report form, a “green sheet.”

USDA Forest Service Gen. Tech. Rep. PSW-GTR-191. 2005 720 Autumn Hawk Watch - McCarty and Bildstein

In order to simplify the transition for hawkwatchers watchsite, shown in tabular format for a specific date from a paper to an electronic archive, the Autumn or over 5-day periods (fig. 3). Hawk Watch data entry web page was designed after the DRF and uses the same codes for weather variables The effectiveness of the Autumn Hawk Watch database already familiar to hawkwatchers. After data entry, toward realizing the potential to track raptor populat- watchsite coordinators can print summaries of daily ions is tied to ensuring accuracy of data entry and help- counts for their site, and users of Autumn Hawk Watch ing project coordinators to control data quality for their can view and print counts from all watchsites, making watchsite. To reduce errors made during data entry, the timely entry of count data by each watchsite coor- CLO and HMS fulfill administrative functions within dinator essential to the value of the web page to other Autumn Hawk Watch, such as maintaining regional users. checklists of migratory raptors, placing count limits on each species based on geographic regions and time of To enhance the educational impact of the raw count year, and alerting users if they report unusual sightings data, and to track the migration as it happens, Autumn of a species or enter counts exceeding the limits. An- Hawk Watch generates a series of animated maps for other important feature of Autumn Hawk Watch is its 29 raptor species showing the abundance of migrants ability to accept sporadic or even anecdotal data. Thus, recorded at each watchsite over 1-day, 5-day, or 10-day single days of observation at an individual site, or even intervals throughout the entire count period, generally partial days of observation, can be accepted and archiv- running from August to December, and also creates ed in the geographically explicit database for future cumulative, static maps and tables by species over 5- use. Since many watchsites are operated strictly on a day or 10-day periods, or for the entire count period part-time or volunteer basis, this type of information is (fig. 2). Users also can view counts from a single important because it builds a base on which to expand further monitoring efforts.

Figure 2—Autumn Hawk Watch map showing counts of Peregrine Falcons (Falco peregrinus) from 1 August to 9 December 2001 at all watchsites.

USDA Forest Service Gen. Tech. Rep. PSW-GTR-191. 2005 721 Autumn Hawk Watch - McCarty and Bildstein

for for for for for for 10-26 to 10- Total count up Species 10-26 10-27 10-28 10-29 10-30 10-31 31 to 10-31 Black Vulture Coragyps atratus 0 0 2 0 2 0 4 28 Cathartes aura 6 20 10 3 32 0 71 240 Osprey Pandion haliaetus 2 1 0 0 0 0 3 573 Bald Eagle Haliaeetus 0 3 1 0 1 0 5 112 leucocephalus Northern Harrier Circus cyaneus 5 18 6 1 2 1 33 194 Sharp-shinned Hawk Accipiter 40 201 25 7 10 2 285 4664 striatus Cooper's Hawk Accipiter cooperii 12 50 17 0 10 1 90 820 Accipiter 0 4 13 0 4 1 22 30 gentilis Red-shouldered Hawk Buteo 1 17 49 16 12 1 96 180 lineatus Broad-winged Hawk Buteo 0 0 0 0 0 0 0 3842 platypterus Red-tailed Hawk Buteo 52 579 521 68 123 3 1346 2158 jamaicensis Aquila chrysaetos 3 5 8 2 1 0 19 35 Falco 3 0 1 0 0 0 4 580 sparverius Falco columbarius 3 2 0 0 0 0 5 170 Falco 0 0 0 0 0 0 0 34 peregrinus raptor sp. Falconiformes 2 9 0 2 0 0 13 121 Total 129 909 653 99 197 9 1996 13781 Figure 3—Table generated by the Autumn Hawk Watch web page showing counts of migrant raptors over a 5-day period (26 to 31 October 2001) and total count for the season for Hawk Mountain Sanctuary.

Results from the first four years of the Autumn Hawk entered the data into the Autumn Hawk Watch database. Watch project indicate that the long-range plans to Data entry by watchsite cooperators increases both the organize hawk migration data, to make the data more accuracy and timely entry of data, and the website accessible, and to improve techniques to monitor raptor works best when participants directly enter their own populations are being accomplished. Overall, 99 dif- data. All count data appearing on the website also are ferent watchsites in five countries in North, Central, published in the HMANA Journal of Hawk Migration and South America entered data into the Autumn Hawk Studies. Watch database in at least one year from 1999 to 2002. Fifty-one watchsites participated in 1999, 53 in 2000, By 2002, count data were reported from three Canadian 58 in 2001, and 66 in 2002. In 2001, 53 watchsites Provinces, 26 United States, Mexico, Costa Rica, and were in North America and five were in Central Amer- Bolivia. Fifty-two percent of the sites were in north- ica (fig. 4; tables 1, 2). The watchsites averaged 57 eastern North America, and more than one-quarter of days of observation (3333 total days) in 2001, mostly all sites were in either Pennsylvania or Connecticut. between 26 August and 1 November. (Although imp- Geographic distribution reflects the numbers of watch- ressive, these numbers represent but a fraction of the sites available. Currently, more participation is needed vast amount of effort expended each year by hawk- from the western and, particularly, the central flyways watchers in North and Central America.) Autumn Hawk of North America, including the entire Mississippi Watch tracked the movements of 28 species of raptors River corridor. in 2001. Most data entry was performed by cooperators at each site; however 16 watchsites provided electronic copies or field sheets of counts to HMS, which then

USDA Forest Service Gen. Tech. Rep. PSW-GTR-191. 2005 722 Autumn Hawk Watch - McCarty and Bildstein

Conclusions science contribution number 95 from Hawk Mountain Sanctuary. The long-term goal of Autumn Hawk Watch is to ad- vance the ability of ornithologists and amateur hawk- Literature Cited watchers to monitor continental populations of raptors efficiently and to allow rapid information transfer. The Bednarz, J. C., D. Klem, L. J. Goodrich, and S. E. Senner. 1990. Migration counts at Hawk Mountain, Pennsylvania, as Autumn Hawk Watch website is designed to accom- indicators of population trends, 1934-1986. Auk 107: 96- modate data for 200 known watchsites in North, Cen- 109. tral, and South America, with the potential to add any number of new watchsites. Currently, data can be Berthold, P. 1993. Bird migration. Oxford: Oxford University entered from 1998 to the present, and planned entry of Press. historical DRFs from the archive will enhance the ex- Bildstein, K. L. 1998. Long-term counts of migrating raptors: panding database. In addition to the benefits to the A role for volunteers in wildlife research. Journal of hawkwatching and scientific communities, Autumn Wildlife Management 62: 435-445. Hawk Watch has the potential for broadening the gen- eral public’s knowledge of and exposure to the timing, Bildstein, K. L., and R. A. Compton. 2000. Mountaintop science: The history of conservation ornithology at geography, and extensiveness of hawk migration in the Hawk Mountain Sanctuary. In: W. E. Davis and J. A. Americas. Jackson, editors. Contributions to the history of North American Ornithology. Vol. II. Cambridge, MA: Nuttall Ornithological Club; 153-182.

Bildstein, K. L. and J. Zalles. 2001. Raptor migration along the Mesoamerican land corridor. In: K. Bildstein and D. Klem, editors. Hawkwatching in the Americas. North Wales, PA: Hawk Migration Association of North America; 119-141.

Carson, R. 1962. Silent Spring. Boston, MA: Houghton-Mifflin.

Dunn, E. H., and D. L. Tessaglia. 1994. Predation of birds at feeders in winter. Journal of Field Ornithology 65: 8-16.

Kerlinger, P. 1989. Flight strategies of migrating hawks. Chicago, IL: University of Chicago Press.

McCarty, K., K. Arnold, J. Ottinger, and K. L. Bildstein. 2000. HMANA data at Hawk Mountain Sanctuary: An update through January 2000. HMANA Hawk Migration Studies 25(2): 30-39.

Mueller, H. C., D. D. Berger, G. Allez, N. S. Mueller, W. G. Robichaud, and J. L. Kaspar. 2001. Migrating raptors and vultures at Cedar Grove, Wisconsin, 1936-1999: An index of population changes. In: K. Bildstein, and D. Klem, editors. Hawkwatching in the Americas. North Wales, PA: Hawk Migration Association of North America; 1-22.

Newton, I. 1979. Population ecology of raptors. Vermillion, SD: Buteo Books.

Roberts, P. M. 2001. In the beginning: The origins and early Figure 4—Watchsite locations years of HMANA. In: K. Bildstein and D. Klem, editors. Hawkwatching in the Americas. North Wales, PA: Hawk Migration Association of North America; 263-268.

Acknowledgments Smith, J. P., and S. W. Hoffman. 2000. The value of extensive Autumn Hawk Watch would not be possible without the raptor migration monitoring in western North America. continued support of coordinators at each watchsite and In: R. D. Chancellor, and B.-U. Meyburg, editors. Raptors at risk. Berlin: World Working Group for Birds of Prey and their associated staff and volunteers, as well as the ef- Owls; 597-615. forts of S. Kelling and his co-workers at the Cornell Lab of Ornithology who oversee the Autumn Hawk Smith, J. P., J. Simon, S. W. Hoffman, and C. Riley. 2001. New Watch web page. We thank them all. This is conservation full-season autumn hawkwatches in coastal Texas. In: K. Bildstein and D. Klem, editors. Hawkwatching in the

USDA Forest Service Gen. Tech. Rep. PSW-GTR-191. 2005 723 Autumn Hawk Watch - McCarty and Bildstein

Americas. North Wales, PA: Hawk Migration Association Zalles, J. I., and K. L. Bildstein, editors. 2000. Raptor watch: A of North America; 67-91. global directory of raptor migration sites. Cambridge, England: Birdlife International; and Kempton, PA: Hawk Viverette, C. B., S. Struve, L. J. Goodrich, and K. L. Bildstein. Mountain Sanctuary. 1996. Decreases in migrating Sharp-shinned Hawks (Accipiter striatus) at traditional watchsites in eastern North America. Auk 113: 32-40.

Table 1—Number of watchsites participating in Autumn Hawk Watch per year. Year United States Canada Mexico Costa Rica Bolivia Total 1999 38 4 2 7 - 51 2000 43 5 2 1 2 53 2001 48 5 2 3 - 58 2002 57 6 1 2 - 66

Table 2—Autumn Hawk Watch participants, 1999-2002.

Continent Continent State, country, State, country, or province Watch site or province Watch site North America: United States Arizona Lipan Point, Grand Canyon Massachusetts Blueberry Hill Yaki Point, Grand Canyon Bolton Flats Mount Watatic Connecticut Bent-of-the-River Sanctuary Wachusett Mountain State Briggs Hill Reservation Chestnut Hill Mohonk Preserve Flat Hill Flirt Hill Michigan Southeastern Michigan Raptor Good Hill Research Heritage Village Huntington State Park Minnesota Hawk Ridge Nature Reserve Johnnycake Mountain Farm Lighthouse Point Montana Bridgers Mountains Maltby Lakes Osborne Hill Nevada Goshute Mountains Quaker Ridge Torrington Middle School New Hampshire Little

Florida Curry Hammocks New Jersey Cape May Point Raptor Idaho Lucky Peak Banding Station Montclair Hawk Lookout Illinois Illinois Beach State Park Picatinny Peak Picatinny Peak – The Domes Iowa Hitchcock Nature Area Raccoon Ridge MacBride Nature Recreation Area Scotts Mountain Louisiana Northern Shore of Lake Calcasieu Sunrise Mountain, Stokes State Forest Maine Acadia National Park Wildcat Ridge Hawkwatch Harpswell Peninsula/Casco Bay New Mexico Manzano Mountains

USDA Forest Service Gen. Tech. Rep. PSW-GTR-191. 2005 724 Autumn Hawk Watch - McCarty and Bildstein

Table 2 (continued). Continent Continent State, country, State, country, or province Watch site or province Watch site North America: United States (contd.) Central Park Pennsylvania Waggoner’s Gap Chestnut Ridge Hawk Watch (contd.)

North Carolina Chambers Mountain South Carolina Caesar’s Head Mahogany Rock Mountain Tibwin Plantation Mount Pisgah Pilot Mountain State Park Texas Hazel Bazemore Hawk Watch Smith Point Oregon Bonney Butte Utah Wellsville Mountains Pennsylvania -Central City Vermont Putney Mountain Brady's Bend Hawk Mountain Sanctuary Virginia Candler Mountain Little Gap Kiptopeke State Park Militia Hill Rockfish Gap Rocky Ridge County Park Snickers Gap Rose Tree Park Second Mountain Washington Chelan Ridge Stone Mountain

North America: Canada British Rocky Point Bird Observatory Ontario (contd.) Holiday Beach Migration Columbia Observatory Iroquois Shoreline Raptor Watch Ontario Cranberry Marsh St.-Anne-de-Beaupré Hawk Cliff High Park Quebec St.-Anne-de-Beaupré

Central and South America Bolivia Concepción Costa Rica Matina Viru-Viru (contd.) Puerto Viejo Sarapiqui Sixaola Costa Rica Bordon Valle La Estrella Colegio de Bribri Curridabat, San Jose Mexico Cardel E.A.R.T.H. Chichicaxtle Kéköldi

USDA Forest Service Gen. Tech. Rep. PSW-GTR-191. 2005 725